Fault diagnosis and data recovery of sensor based on relevance vector machine

Aiming at the limitation of the fault condition can not be evaluated and confirmed by the traditional sensor, the fault diagnosis and data recovery methods based on relevance vector machine are researched in this paper. Because the fault type of sensor is not only one kind, it solves fault diagnosis by multiple classifiers based on relevance vector machine; it achieves data recovery of sensor by regression principle based on relevance vector machine, which uses the normal output data before the fault. Finally, the pressure sensor is analyzed, the results indicates that, it is effective to diagnose the fault state of sensor, and the data recovery is achieved using this method. It promoted the reliability of sensor.